sm: sm.ts.pdf – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

sm.ts.pdf

Nonparametric density estimation of stationary time series data

Description

This function estimates the density function of a time series x, assumed to be stationary. The univariate marginal density is estimated in all cases; bivariate densities of pairs of lagged values are estimated depending on the parameter lags.

Usage

sm.ts.pdf(x, h = hnorm(x), lags, maxlag = 1, ask = TRUE)

Arguments

`x`	a vector containing a time series
`h`	bandwidth
`lags`	for each value, `k` say, in the vector `lags` a density estimate is produced of the joint distribution of the pair `(x(t-k),x(t))`.
`maxlag`	if `lags` is not given, it is assigned the value `1:maxlag` (default=1).
`ask`	if `ask=TRUE`, the program pauses after each plot, until <Enter> is pressed.

Details

see Section 7.2 of the reference below.

Value

a list of two elements, containing the outcome of the estimation of the marginal density and the last bivariate density, as produced by sm.density.

Side Effects

plots are produced on the current graphical device.

References

Bowman, A.W. and Azzalini, A. (1997). Applied Smoothing Techniques for Data Analysis: the Kernel Approach with S-Plus Illustrations. Oxford University Press, Oxford.

Examples

with(geyser, {
   sm.ts.pdf(geyser$duration, lags=1:2)
})

sm

Smoothing Methods for Nonparametric Regression and Density Estimation

v2.2-5.6

GPL (>= 2)

Authors

Adrian Bowman and Adelchi Azzalini. Ported to R by B. D. Ripley <ripley@stats.ox.ac.uk> up to version 2.0, version 2.1 by Adrian Bowman and Adelchi Azzalini, version 2.2 by Adrian Bowman.

Initial release

2018-09-27